On Mon, Mar 20, 2006 at 08:28:13PM -0500, Jeff Garzik wrote: > Thanks a lot for testing. > > I've stored the sata_nv updates I sent you in the 'nv-adma' branch of > git://git.kernel.org/pub/scm/linux/kernel/git/jgarzik/libata-dev.git OK, I'll pull from there for further testing. > Dumb question, to be certain that I understood your first paragraph: > you do indeed see at least -some- success talking to devices on port 1, > 2, 3... ? i.e. not just port 0 works? I can start up the RAID1's on ports 2 and 3, activate the VG on top of /dev/md5, mount the filesystems, and use them. I was able to cp -a the mounted root filesystem from /dev/md2 to a subdir on /dev/sda1. It was only when I started up a few copies of "tar cf /dev/zero ... &" that the timeouts began. > Weird. Well, now that we appear to have narrowed the non-ADMA case down > to inb(), I'm going to punt this one as not-my-problem ;-) Roger. I'll stare at it a bit longer, but I probably need to order three SATA adapters. Any recommendations? These Tyan 2895 machines all have 2-4 74GB WD Raptors in them. I don't really need expensive 3ware or Areca cards for these workstations, just reasonable latencies. Thanks. -Bill - : send the line "unsubscribe linux-ide" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
